Methods were developed for quantifying the source fluxes of the excitotoxin quinolinic acid (QUIN) in normal and inflamed tissue. The issue of sources of QUIN is important, because potential therapeutic strategies to reduce QUIN accumulations and neurotoxicity may target different potential sources. We developed an isotope ratio assay in which unlabeled QUIN:labeled QUIN concentration ratios, for individual tissues and plasma in animals infused to steady state with labeled QUIN, could be related to the fraction f of QUIN produced endogenously, in relation to that obtained from all sources. We performed experiments on control, cerebral ischemic, and LPS-inoculated gerbils and rats infused with labeled QUIN via osmotic minipumps. We employed a highly sensitive mass spectrometry assay for determination of labeled and unlabeled QUIN levels in brain and other tissues, plasma, whole blood, and urine. A major finding is that endogenous production of QUIN is the dominant source of this agent during brain inflammation, since nearly all hippocampus and striatal QUIN is synthesized endogenously at peak response to ischemia (f=95%); even 30% of QUIN present in control brain is synthesized endogenously. This finding implies that agents that act at the terminal step of QUIN synthesis, such as 6-chlorohydroxyanthranilate, should be effective in reducing QUIN production and related toxicity during brain inflammation. In addition, microdialysis has been introduced to determine absolute fluxes of QUIN production in tissue. Modeling of the kynurenine pathway combined with microdialysis has shown that depletion of QUIN precursors by dialysis does not significantly alter the rate of production of QUIN in brain tissue.